Vitamin A supplementation boosts control of antibiotic-resistant Salmonella infection in malnourished mice

Abstract

Disseminated disease from non-typhoidal Salmonella enterica strains results in >20% mortality globally. Barriers to effective treatment include emerging multidrug resistance, antibiotic treatment failure, and risk factors such as malnutrition and related micronutrient deficiencies. Individuals in sub-Saharan Africa are disproportionately affected by non-typhoidal S. enterica bloodstream infections. To inform a clinical trial in people, we investigated vitamin A as a treatment in the context of antibiotic treatment failure in a mouse model of vitamin A deficiency. Vitamin A-deficient (VAD) mice exhibited higher systemic bacterial levels with a multidrug-resistant clinical isolate in comparison to mice on a control diet. Sex-specific differences in vitamin A deficiency and disseminated infection with S. enterica serotype Typhimurium (S. Typhimurium) were observed. VAD male mice had decreased weight gain compared to control male mice. Further, infected VAD male mice had significant weight loss and decreased survival during the course of infection. These differences were not apparent in female mice. In a model of disseminated S. Typhimurium infection and antibiotic treatment failure, we assessed the potential of two consecutive doses of vitamin A in alleviating infection in male and female mice on a VAD or control diet. We found that subtherapeutic antibiotic treatment synergized with vitamin A treatment in infected VAD male mice, significantly decreasing systemic bacterial levels, mitigating weight loss and improving survival. These results suggest that assessing vitamin A as a therapy during bacteremia in malnourished patients may lead to improved health outcomes in a subset of patients, especially in the context of antibiotic treatment failure.

Fig 1. Systemic burden of Salmonella is higher in vitamin A-deficient mice.

To generate vitamin A-deficient (VAD) mice, C57BL/6 Slc11a1^+/+ pregnant mice were put on a VAD diet 2 weeks into gestation (A). At 3 weeks, pups were weaned onto either a VAD or control diet. Male (n = 7) and female (n = 7–8) mice on each diet were infected with S. Typhimurium (STm) D23580 via the intraperitoneal (i.p.) route at 9 weeks of age and systemic bacterial burden was characterized by colony-forming units (CFU) in liver, spleen and blood collected at necropsy (Nx) 4 days after infection (B). Data represent mean ± SEM. Significance between control and VAD mice was determined with a Mann-Whitney test of log-transformed values. A p<0.05 was considered significant.

Fig 2. Increased weight loss and decreased survival seen in VAD male mice infected with Salmonella.

Weight (g) of VAD and control male and female mice was recorded weekly (A) during growth from 4 weeks to 9 weeks (n = 26–30). Nine week-old mice were infected with S. Typhimurium D23580 via the intraperitoneal (i.p.) route and weight (g) was monitored daily (B) for up to 4 days post-infection (n = 8–9). Data are represented as mean ± SEM. Significance between weights by diet or day was determined using a Mann-Whitney test (*, p<0.05; ns, not significant). (C) Percent survival was assessed for all groups over the 4-day infection with a Kaplan-Meier survival curve (n = 8–9). Results are reported as percent survival. Pairwise comparisons between the VAD male group and each other group was determined with a survival analysis log-rank (Mantel-Cox) test (*, p<0.05). All data represent a compilation of data from three independent experiments.

Fig 3. Salmonella strains colonize systemically at higher levels in VAD male mice early in infection.

Systemic bacterial burden was characterized by CFU in liver, spleen and blood collected at necropsy 1 day after infection with S. Typhimurium D23580, SL1344 or SARA16 via the i.p. route for male and female mice (n = 4–8) on either control (A) or VAD (B) diet. Data represent mean ± SEM. Significance between male and female mice was determined with a Mann-Whitney test of log-transformed values (*, p<0.05; ns, not significant).

Fig 4. Antibiotic treatment failure occurs at specific concentrations in vivo.

Enrofloxacin was administered directly into the drinking water on day 2 post-infection with multidrug-resistant S. Typhimurium D23580. Groups of C57BL/6J male (n = 5) and female (n = 5) mice on a standard diet were assessed for the following treatment groups: 0 mg/ml, 0.01 mg/ml, 0.05 mg/ml, and 0.10 mg/ml. Systemic bacterial burden was characterized by CFU in liver, spleen and blood collected at necropsy 5 days after infection (A). Water weights and mouse weights were collected daily. The average enrofloxacin consumed (μg/kg/day) was determined for each group based on water and mouse weights for days 2–3, 3–4 and 4–5 (B). Data are reported as mean ± SEM. Significance was determined on log-transformed values (A) or calculated values (B) with a Kruskal-Wallis test and Dunn’s multiple comparisons test (*, p<0.05; ns, not significant).

Fig 5. Co-treatment of vitamin A and antibiotics decreases systemic Salmonella burden and improves survival in VAD male mice.

VAD and control Slc11a1^+/+ mice were generated and infected with multidrug resistant S. Typhimurium D23580 at 9 weeks of age. Mock-treated mice received 100 μl of sterile PBS administered by oral gavage on day 1 and day 2 post-infection. Vitamin A-treated mice received 600 IU of retinyl palmitate in 100 μl of sterile PBS administered by oral gavage on day 1 and day 2 post-infection and changed to control diet at day 1 post-infection. The antibiotic enrofloxacin (abx) was administered at day 2 post-infection in the drinking water at 0.05 mg/ml (A). Systemic burden of S. Typhimurium was characterized for VAD male (B) and VAD female (C) mice (n = 4–6) as CFU in liver, spleen and blood at necropsy 4 days post-infection. Data are reported as mean ± SEM. Significance was determined on log-transformed values with a Kruskal-Wallis test and Dunn’s multiple comparisons test (*, p<0.05; ns, not significant). (D) Percent survival for control and treated VAD male mice was assessed with an 8-day Kaplan-Meier survival curve (n = 6–16). An enrofloxacin concentration of 0.01 mg/ml was used in drinking water treatment, and oral gavage treatments were given as previously described. Significance was determined with pairwise comparisons between the control group and each other treatment group using a log-rank (Mantel-Cox) test (*, p<0.05; ns, not significant). Data from untreated controls are also included in Fig 1.